brain sciences Article Glutamatergic Activation of Neuronostatin Neurons in the Periventricular Nucleus of the Hypothalamus Sema Serter Kocoglu 1,*, Duygu Gok Yurtseven 2, Cihan Cakir 3, Zehra Minbay 3 and Ozhan Eyigor 3 1 Department of Histology and Embryology, Balikesir University School of Medicine, Balikesir 10145, Turkey 2 Department of Histology and Embryology, Sanko University School of Medicine, Gaziantep 27090, Turkey; dgok@uludag.edu.tr 3 Department of Histology and Embryology, Bursa Uludag University School of Medicine, Bursa 16240, Turkey; cihancakir10@gmail.com (C.C.); zminbay@uludag.edu.tr (Z.M.); oeyigor@uludag.edu.tr (O.E.) * Correspondence: serter_bio@hotmail.com  Received: 27 March 2020; Accepted: 3 April 2020; Published: 6 April 2020  Abstract: Neuronostatin, a newly identified anorexigenic peptide, is present in the central nervous system. We tested the hypothesis that neuronostatin neurons are activated by feeding as a peripheral factor and that the glutamatergic system has regulatory influences on neuronostatin neurons. The first set of experiments analyzed the activation of neuronostatin neurons by refeeding as a physiological stimulus and the effectiveness of the glutamatergic system on this physiological stimulation. The subjects were randomly divided into three groups: the fasting group, refeeding group, and 6-cyano-7-nitroquinoxaline-2,3-dione (CNQX)+refeeding group. We found that refeeding increased the phosphorylated signal transducers and transcription activator-5 (pSTAT5) expression in neuronostatin-positive neurons and that the CNQX injection significantly suppressed the number of pSTAT5-expressing neuronostatin neurons. The second set of experiments analyzed the activation pathways of neuronostatin neurons and the regulating effects of the glutamatergic system on neuronostatin neurons. The animals received intraperitoneal injections of glutamate receptor agonists (kainic acid, α-amino-3-hydroxy-5methyl-4-isoazepropionic acid (AMPA), and N-methyl-D-aspartate (NMDA)) or 0.9% NaCl. The number of c-Fos-expressing neuronostatin neurons significantly increased following the AMPA and NMDA injections. In conclusion, we found that the neuronostatin neurons were activated by peripheral or central signals, including food intake and/or glutamatergic innervation, and that the glutamate receptors played an important role in this activation. Keywords: neuronostatin; Glutamate; c-Fos 1. Introduction Neuronostatin (NST), encoded by the somatostatin gene, is an anorexigenic peptide with 16 amino acids [1]. Immunohistochemical studies have shown that the bodies of neuronostatin-positive neurons are prominent in the hypothalamic anterior periventricular and suprachiasmatic nuclei (SCN), while those of neuronostatin-positive axon terminations are prominent in the arcuate nucleus and median eminence. Also, there are fewer and less densely-marked neuronostatin-expressing cells in the polymorphic layer of the dentate gyrus and motor cortex, amygdala, and cerebellum [2]. The presence of neuronostatin neurons and neuronostatin-positive axon terminations in the areas of the hypothalamus that control food intake, the demonstration of decreased water and food intake in animals following intraventricular neuronostatin administration, and the decreasing dose-dependent and short-term food intake in mice after intraperitoneal neuronostatin administration all suggest that neuronostatin may play a role in the control of appetite and metabolism [3–5]. Brain Sci. 2020, 10, 217; doi:10.3390/brainsci10040217 www.mdpi.com/journal/brainsci Brain Sci. 2020, 10, 217 2 of 13 In the central nervous system, glutamate is the major excitatory amino acid neurotransmitter and its importance in the regulation of the neuroendocrine systems and the hypothalamus-pituitary-endocrine system axis is well documented [6,7]. Glutamate-mediated neurotransmission occurs via G-protein- mediated metabotropic and ion channel-forming ionotropic glutamate receptors. These receptors are classified according to their agonists: N-methyl-D-aspartate (NMDA), α-amino-3-hydroxy-5methyl-4- isoazepropionic acid (AMPA), and 2-carboxy-3-carboxymethyl-4-isopropenylpyrrolidine (Kainate) receptors [8]. One of the methods used to demonstrate glutamatergic innervation in target neurons is to label some specific molecules involved in neurotransmission with different techniques. Glutamate synthesized in the soma is transported to the axon terminals through vesicular glutamate transporters (VGLUT). VGLUTs are molecules specific to the glutamatergic system, and they are used as specific markers of neurons that use glutamate as neurotransmitters in immunohistochemical techniques [9]. However, immunohistochemically, they can only be shown in the terminal boutons of the axon [10]. Therefore, the presence of VGLUTs on the boutons of the target neuron indicates that the neuron receives glutamatergic innervation [11]. In addition, the demonstration of glutamate receptor subunit proteins that mediate the glutamate effect in the target neuron is another indicator of glutamatergic innervation. However, the presence of receptors does not give information about the functionality of the receptors. Therefore, the response of the target neuron to glutamate should be evaluated. For this purpose, animals are administered only glutamate receptor subunit agonists or antagonists/agonists peripherally or centrally, and then changes in the expression of neuronal activation markers in the neuron are determined using immunohistochemical methods. Transcription factors involved in the regulation of gene expression are specific proteins that play an active role in neuronal functioning, neurogenesis, and neuronal innervation. The transcription factors can be activated in the different intracellular signal transduction pathways, or they can be directly activated by ligands, such as glucocorticoids and some vitamins [12–14]. Certain transcription factors are activated by phosphorylation. These transcription factors are phosphorylated in the cytoplasm and translocate to the nucleus to initiate transcription. Immunohistochemically, the expression of proteins in the nucleus is used as a marker of neuronal activation. Transcription factors, such as c-Fos, phosphorylated cAMP response element-binding protein (pCREB), or phosphorylated signal transducers and transcription activators (pSTATs) have been used as markers for determining neuronal activity changes [15–18]. Despite the many publications describing the mechanisms of action of neuronostatin, there is no data about the peripheral and central control systems that play a role in the regulation of neuronostatin neuron activation. Therefore, we hypothesized that neuronostatin neurons are activated by feeding as a peripheral factor and the glutamatergic system has regulatory influences on neuronostatin neurons. The first set of experiments analyzed the activation of neuronostatin neurons by refeeding as a physiological stimulus and the effectiveness of the glutamatergic system on this physiological stimulation. The second set of experiments analyzed the regulating effects of the glutamatergic system on neuronostatin neurons. 2. Materials and Methods 2.1. Animals All animal experiments were performed according to the National Institute of Health Guide for the Care and Use of Laboratory Animals. The experimental procedures were approved by the Experimental Ethical Committee of Bursa Uludag University (Approval No: 2016–4/4). 60-day-old male Sprague-Dawley rats (200–250 g) (n = 45, male), obtained from the Bursa Uludag University Laboratory Animal Breeding, Usage and Research Center, were used in this study. Animals were housed in this center where the light cycle and the temperature were controlled (a 12:12 h light–dark cycle with the lights off at 7:00 am at 21 ◦C) with freely available water). Brain Sci. 2020, 10, 217 3 of 13 2.2. Experimental Groups Experiment 1: Investigation of the effect of refeeding as a physiological stimulus on neuronostatin neurons and the effectiveness of the glutamatergic system. For this purpose, three experimental groups were designed consisting of male rats, with n = 5 per group. After a 48-h fasting period; the refeeding group was allowed to eat ad libitum for 2 h, while the fasting group was unfed and the antagonist group was injected intraperitoneally with non-NMDA glutamate antagonist 6-cyano-7-nitroquinoxaline-2,3-dione (CNQX) 15 min before the 2-h refeeding period (2 mg/kg CNQX in 300 µL distilled water). After the 48-h fasting, refeeding was started at 9:00 am at the beginning of the dark period of the dark–light cycle and the animals were allowed to feed for 2 h. The CNQX injections were performed at 9:00 am. Experiment 2: Investigation of the regulating effects of the glutamatergic system on neuronostatin neurons. Six groups were set up to determine the effect of glutamate on neuronostatin neurons (n = 5 per group). The first group was designated as the kainic acid group. The rats in this group were intraperitoneally injected with kainic acid (2.5 mg/kg in 300 µL distilled water, DW). The kainic acid control group received saline injections (300 µL saline, intraperitoneal, ip). AMPA was administrated to the rats of the second experimental group (AMPA group, 5 mg/kg in 750 µL DW, ip). The AMPA control group received 750 µL of saline (ip). The third group of rats, the NMDA group, was given NMDA (100 mg/kg in 2 mL DW), and the NMDA control group was injected with 2 mL saline (ip). All drugs were administered between 9:00 a.m. and 11:00 a.m. Ninety minutes after injection, deep anesthesia was performed using ether, and the animals were sacrificed by transcardial perfusion with 4% paraformaldehyde (PFA) in 0.13 M Sorenson’s phosphate buffer, pH 7.4 (300 mL/animal). All brains were removed and post-fixed at +4 ◦C overnight in 4% paraformaldehyde. Five series of brain sections with a thickness of 40 µm were cut using a vibratome. The sections along the rostral-caudal axis of the hypothalamus were collected in Tris-HCl buffer (0.05 M, pH 7.6). The brain sections were then washed in buffer and stored at −20 ◦C in cryoprotectant until use. 2.3. Immunohistochemistry Between incubations, we washed the sections in Tris-HCl three times. A blocking buffer was used to dilute the primary and the secondary antibodies. Blocking buffer was prepared in Tris-HCl buffer containing normal horse serum (10%), Triton X-100 (0.2%), and sodium azide (0.1%). Before the primary antibody incubation, the sections were exposed to blocking buffer for 2 h for the prevention of non-specific binding. The sections were brought to room temperature and the cryoprotectant was removed by buffer washes. Free floating sections were then incubated in preheated antigen retrieval (AR) solution (final solution temperature 73–75 ◦C) for 30 min. Trisodium citrate buffer (50 mM, pH 6, for neuronostatin) or ethylenediaminetetraacetic acid (EDTA) solution (1 mM, pH 8, for phosphorylated signal transducers and transcription activator-5 (pSTAT5)) was used in the AR process. Following the AR, endogenous peroxidase activity was stopped using 3% H2O2 and then the sections were treated with blocking buffer. Primary antibody incubation for c-Fos (rabbit anti-c-Fos, 1/20,000, Chemicon, Billerica, MA, USA) was carried out overnight at room temperature. For pSTAT5 immunohistochemistry, rabbit anti-pSTAT5 antibody (1/2000 dilution, 9351, Cell Signaling Technology, Danvers, MA, USA) was used for 6 nights at +4 ◦C. After the primary antibody incubation, the sections were transferred into the secondary antibody solution containing biotin conjugated donkey anti-rabbit immunoglobulin G (IgG, 1/300, Jackson Immunoresearch Labs, West Grove, PA, USA) and incubated for 2 h. The sections were then washed three times and exposed to avidin-biotin complex (ABC Elite Standard Kit, Vector Labs, Burlingame, CA, USA). Diaminobenzidine (DAB) solution (25 mg DAB, 2 g nickel ammonium sulfate, 2.5 µL hydrogen peroxide in 100 ml Tris-HCl buffer) was used in order to make the immunoreactive signal visible under the microscope. Brain Sci. 2020, 10, 217 4 of 13 For the second primary antibody detection of the double immunohistochemistry, the sections were blocked and incubated in rabbit anti-neuronostatin antibody (1/4000 dilution, H-060-50, Phoenix Pharmaceuticals, Inc., Burlingame, CA, USA) for three nights at room temperature. A secondary antibody (biotin conjugated donkey anti-rabbit IgG (1/400, Jackson Immunoresearch Labs, West Grove, PA, USA) incubation step was performed for 2 h after the primary antibody incubation. Sections were then exposed to a avidin-biotin complex and DAB solution (25 mg DAB, 2.5 µL hydrogen peroxide in 50 mL Tris-HCl buffer). At the end of the double-immunostaining, the sections were mounted on slides, dried, and coverslipped with DPX. For the negative control experiments primary antibody or secondary antibody steps were omitted. 2.4. Statistical Analysis Sections were analyzed and the images were captured with an Olympus BX-50 photomicroscope attached to a charge-coupled device (CCD) camera (Olympus DP71, 1.5 million pixels, Olympus Corporation, Tokyo, Japan). Sections between the coordinates (bregma-0.24 mm to −3.60 mm for the periventricular nucleus), determined according to the rat brain atlas [19], were used for single and double immunohistochemical labeling. We used cross sections taken at five different levels at the same coordinate and at equal distance for each animal in the rostrocaudal plane for cell counting. In dual indirect immunoperoxidase-labeled sections, the ratio of c-Fos- or pSTAT5-expressing neurons to all neuronostatin neurons was calculated for each animal. Counts were performed ‘blindly’ and by two different researchers, and all neuronostatin neurons in the periventricular nucleus were counted. The intra-group mean and standard error of the mean (SEM) of the percentages obtained for each subject were determined. The analysis of variance between the experimental groups was statistically compared using a one-way ANOVA test, and p < 0.05 was considered significant. 3. Results 3.1. Neuronostatin Neuron Immunoreactivity in the Hypothalamus Neuronostatin immunoreactive neuron bodies were expressed in the anterior hypothalamic periventricular nucleus around the third ventricle (Figure 1). Neuronostatin-positive axon terminals were localized in the suprachiasmatic nucleus (Figure 2A), ventromedial hypothalamic nucleus with arcuate nucleus (Figure 2B), and median eminence (Figure 2C). 3.2. Physiological Stimulation of Neuronostatin Neurons and Investigation of the Effects of Glutamate Antagonists We found that 2-h refeeding after 48-h fasting induced phosphorylated signal transducers and transcription activator-5 (pSTAT5) expression in neuronostatin neurons of the anterior hypothalamic periventricular nucleus. The 2-h refeeding after 48-h fasting caused a significant increase in the number of pSTAT5-positive neuronostatin neurons when compared with the fasting group (p < 0.001, Figure 3). The injection of specific antagonist CNQX prior to agonist significantly decreased the percentage of pSTAT5-positive neuronostatin neurons (p < 0.01, Figure 3). Brain Sci. 2020, 10, 217 5 of 13 Figure 1. Distribution of neuronostatin neurons in the hypothalamus. (A,C,D) Neuronostatin neurons located around the third ventricle in the periventricular nucleus of the anterior hypothalamus. (B) Neuronostatin-positive axonal outcomes observed in ventromedial hypothalamic nucleus (VMH). 3V: third ventricle, OK: optic chiasm. Figure 2. Distribution of neuronostatin-positive axon terminals in the hypothalamus. (A) Suprachiasmatic nucleus (SCN), (B) ventromedial hypothalamic nucleus (VMH) and arcuate nucleus (ARC), and (C) median eminence. 3V: third ventricle, OK: optic chiasm. Brain Sci. 2020, 10, 217 6 of 13 Figure 3. The effect of CNQX, a non- N-methyl-D-aspartate (NMDA) glutamate antagonist, in subjects Brain Sci. 2020, 10, x FOR PEER REVIEW stimulated by refeeding after fasting. There were significant diff7e oref n14c es between fasting-refeeding groups (*** p < 0.001) and refeeding-antagonist groups (** p < 0.01) (B). pSTAT5 expression in neuronostatin neurons after 48 h fansetuinrgo n(oosnt athtien lneeftu),r opnSsTAafTte5r e4x8phrefsassiotinn gin( onneuthroenloefstt)a,tpinS TnAeuTr5oenxsp ression in neuronostatin neurons activated with refeeding in antaecrtioivra thedypwoitthharleafmeeidc inpgeirnivaenntetrriicourlhayr pontuhcalleaums ic(pine rivtheen trmiciudladrlen)u, cleus (in the middle), neuronostatin neuronostatin neurons in the antneeriuorro npseirnivtehnetrainctuelrairo rnpuecrlievuesn tirnic uClaNrQnXu-ctlreeuasteind CsNubQjeXc-ttsr ebaetefodrseu bjects before feeding (on the right) feeding (on the right) (A). pSTAT(A5 )p. proStTeAinT-5expproretesisnin-egx pnreeussrionngonsteautrinon noestuartoinnsn e(u▲ro) nasn(dN )paSnTdApTST5A- T5-negative neuronostatin neurons negative neuronostatin neurons ( ( ) . 3 V).: 3tVhi:r tdhivredn vtreinclter.icle. 3.3. Investigation of the Glutamatergic SInystehme Erefffeecetds ianngd Agrcotiuvpat,ioanb Pouathw25a.y5s6 in± th2e. 5R2e%guolaftiothne ofn euronostatin neurons localized in the Neuronostatin Neurons periventricular zone were pSTAT5-positive, whereas this ratio was 8.18 ± 0.87% in the fasting group. The ratio of the activated neuronostatin neurons was reduced to 15.08 ± 1.73% after CNQX injection In the anterior hypothalamic periventricular nucleus, c-Fos expression in neuronostatin neurons (Figure 3). was detected following the glutamate agonist administration. The greatest increase in the number of c-Fos expressing neuronostatin 3n.e3u. rIonnvess wtigaast icoanuosfetdh ebyG ltuhtea mNaMteDrgAic iSnyjsetcetmioEn.ff e cts and Activation Pathways in the Regulation of In the kainic acid group, thNeerue rwonaoss tnaoti nsiNgneuifriocnans t increase (0.78 ± 0.51% vs. 0.88 ± 0.55%) in the number of c-Fos-positive neuronostatin neurons when compared to the control group (Figure 4). The In the anterior hypothalamic periventricular nucleus, c-Fos expression in neuronostatin neurons number of c-Fos-positive neuronostatin neurons was significantly increased by AMPA (11.31 ± 1.56% was detected following the glutamate agonist administration. The greatest increase in the number of vs. 4.6 ± 0.86%, p < 0.01) (Figure 5) and NMDA (15.74 ± 1.80% vs. 2.37 ± 0.58%, p < 0.001) (Figure 6) in c-Fos expressing neuronostatin neurons was caused by the NMDA injection. comparison to the control group. In the kainic acid group, there was no significant increase (0.78 ± 0.51% vs. 0.88 ± 0.55%) in the number of c-Fos-positive neuronostatin neurons when compared to the control group (Figure 4). The 7 Brain Sci. 2020, 10, 217 7 of 13 number of c-Fos-positive neuronostatin neurons was significantly increased by AMPA (11.31 ± 1.56% vs. 4.6 ± 0.86%, p < 0.01) (Figure 5) and NMDA (15.74 ± 1.80% vs. 2.37 ± 0.58%, p < 0.001) (Figure 6) in comparison to the control group. Brain Sci. 2020, 10, x FOR PEER REVIEW 7 of 14 neuronostatin neurons after 48 h fasting (on the left), pSTAT5 expression in neuronostatin neurons activated with refeeding in anterior hypothalamic periventricular nucleus (in the middle), Figure 4. Effneeuctroonf oksatiantiicna cnieduardonmsi niins trtahtei oannotnercio-Fr opseerxipvreenstsriiocnulianr nneuucrloenuoss tiant inCnNeQuXro-ntrsealotecadt esduibnjects before the anteriorfeheydpiontgh a(loanm itchep erriigvhent)t r(iAcu).l apr SnTuAclTeu5 sp(Aro)t.eci-nF-oesxpprreostseiinng-e xnperuersosinnogstnaetuinr onneoustraotnins n(▲eu)r oannsd pSTAT5- (N) and c-Fonse-gnaetgivatei vneeunreounroonstoasttiant inneunerounros n( s ( ) . N )o. 3sVig:n tihfiicrdan vtednitffreicrleen. ces were found between the control and kainic acid groups (B). 3V: third ventricle. 3.3. Investigation of the Glutamatergic System Effects and Activation Pathways in the Regulation of Neuronostatin Neurons In the anterior hypothalamic periventricular nucleus, c-Fos expression in neuronostatin neurons was detected following the glutamate agonist administration. The greatest increase in the number of c-Fos expressing neuronostatin neurons was caused by the NMDA injection. In the kainic acid group, there was no significant increase (0.78 ± 0.51% vs. 0.88 ± 0.55%) in the number of c-Fos-positive neuronostatin neurons when compared to the control group (Figure 4). The number of c-Fos-positive neuronostatin neurons was significantly increased by AMPA (11.31 ± 1.56% vs. 4.6 ± 0.86%, p < 0.01) (Figure 5) and NMDA (15.74 ± 1.80% vs. 2.37 ± 0.58%, p < 0.001) (Figure 6) in comparison to the control group. 7 Brain Sci. 2020, 10, 217 8 of 13 Brain Sci. 2020, 10, x FOR PEER REVIEW 7 of 14 neuronostatin neurons after 48 h fasting (on the left), pSTAT5 expression in neuronostatin neurons activated with refeeding in anterior hypothalamic periventricular nucleus (in the middle), Figure 5n.eTuhroeneoffsetacttionf AneMuProAnsa dimn inthiset raantitoenrioonr cp-eFroisveenxptrriecsusliaorn ninucnleeuurso nino sCtaNtinQXne-turreoantesdl oscuatbejdecitns before the antefreieodr ihnygp o(othna ltahme ircigphert)iv (eAnt)r. icpuSlTaAr nTu5c lperuost.ecin-F-eoxspprreostseiinng-e xnperuersosninogstnaetiunr onneousrtoantisn n(▲eu)r oannsd( Np)STAT5- and c-Fonse-gnaetgivaeti vneeunreounroosntoasttinat inneunreounros n( s ( ) . c -)F. o3sVe: xthpirredss vioenntirnicnlee.u ronostatin neurons localized in the anterior hypothalamic periventricular nucleus (A). Significant differences were found between the co3n.t3ro. lInanvedstAigMatPiAong orof uthpes G(*l*upta